I've run across a rather nasty bug in 8.1.2. It seems when the
planer uses an index_scan within a GroupAggregate for a multi column
index you can get incorrect results. fwiw i also see this on a dual
xeon box running 8.1.1 and redhat 7.3.
I've created a simple test case that I hope isola
"Nicholas" <[EMAIL PROTECTED]> writes:
> SELECT cgm.id, g.name, mi.invitees FROM meeting_instances as mi JOIN
> meeting_versions as m ON m.iid = '33' JOIN groups as g ON g.id = m.team LEFT
> JOIN cachedgroupmembers as cgm ON cgm.groupid = m.team and not cgm.disabled
> and cgm.memberid = '55' WHERE
The following bug has been logged online:
Bug reference: 2219
Logged by: Ricardo Solanilla
Email address: [EMAIL PROTECTED]
PostgreSQL version: 8.0
Operating system: Windows XP
Description:bug in 12000 rows update
Details:
i has report this bug earlier (ref. 2210) b
The following bug has been logged online:
Bug reference: 2218
Logged by: Nicholas
Email address: [EMAIL PROTECTED]
PostgreSQL version: 8.1.2
Operating system: Gentoo Linux x86
Description:Variables selected in VIEWs under different names break
queries using those view
Peter Eisentraut <[EMAIL PROTECTED]> writes:
> This bug was reported to Debian. Comments?
These are all known issues. The implementation of domains really needs
rather a lot of work :-(.
One thing we might think about is having domain types use a special set
of input functions rather than just
This bug was reported to Debian. Comments?
Tim Southerwood wrote:
> Package: postgresql-8.1
> Version: 8.1.2-1
> Severity: important
>
> DOMAIN CHECK constraint is bypassable when inserting rows using
> perl/DBD::Pg AND prepare/execute semantics AND using bind values.
>
> This is serious as data
